crash Rstudio in AWS (linux based) after using a function

General
#1

I developed my R code on my local computer. In my code, I mainly use swmmr package to run SWMM (stormwater management model) and further analysis related to the model outputs. To do that, I have used multiple packages (dplyr, shiny, leaflet, xts, zoo, ...). Everything is fine when I use my local computer and I can run my code from first to last and it works properly.

Now, I decided to move all my codes to AWS EC2 cloud. Because I need to run my code every one hour. I had some difficulties with this transformation, but I could solve most of them.

However, somewhere in my code, when I use the read_out() function from swmmr package, I get some warnings about masking some functions from other packages (attached figure) and my studio in AWS server crash.

Why does RStudio in AWS crash when I use this function? All other functions in this package works for me, just this function which I really need to use is not working and making the crash in Rstudio. As I said, it works properly in Rstudio on my local computer.
Any idea about fixing this problem would be appreciated.

Capture3
Capture31159×767 78.3 KB

#2

Hi, welcome!

We don't really have enough info to help you out. Could you ask this with a minimal REPRoducible EXample (reprex)? A reprex makes it much easier for others to understand your issue and figure out how to help.

If you've never heard of a reprex before, you might want to start by reading this FAQ:

Also, please notice that RStudio server running on AWS is not the same as RStudio Cloud which is a cloud service offered by RSTUDIO, not by AWS

#6

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

#4

I suspect you are running out of RAM memory in your EC2 instance, I have tried to reproduce on AWS but I get a different error

library(swmmr)

download.file("https://github.com/HamedGhodsi90/CSO_Real-Time_Prediction/blob/master/SWMM_online.out",
              "SWMM_online.out")

results <- read_out(file = "SWMM_online.out", iType = 1, object_name = "10058", vIndex = 4)
#> Warning in read_out(file = "SWMM_online.out", iType = 1, object_name =
#> "10058", : error reading out file

results
#> $error
#> [1] 1

Created on 2020-06-25 by the reprex package (v0.3.0)

#3

Thank you for your replay.
So if you wanna replicate my work, you can download the .out file from this link:
https://github.com/HamedGhodsi90/CSO_Real-Time_Prediction/blob/master/SWMM_online.out

Then, use this code:

#install.packages("swmmr")
library(swmmr)
results <- read_out(file = "C:/ThePathToFile/SWMM_online.out", iType = 1, object_name = "10058", vIndex = 4)

results

if you try it in your local Rstudio, it should work and the result would be like this:
`10058` 10058$total_inflow
[,1]
2020-06-22 11:09:00 0
2020-06-22 11:24:00 0
2020-06-22 11:39:00 0
2020-06-22 11:54:00 0
2020-06-22 12:09:00 0
2020-06-22 12:24:00 0
2020-06-22 12:39:00 0
2020-06-22 12:54:00 0

But when I use AWS (amazon web services) ec2 and run this model in its studio, I get the error (the figure above) that says "the previous R session was abnormally terminated due to an unexpected crash".

Is it enough for replicating my code?

#5

Thank you for your reply. I believe the source file I uploaded was changed. I uploaded it again.
Can you run your model again to see if your Rstudio crash or not.
use this link plz:
https://github.com/HamedGhodsi90/CSO_Real-Time_Prediction/raw/master/SWMM_online.out

download.file("https://github.com/HamedGhodsi90/CSO_Real-Time_Prediction/raw/master/SWMM_online.out", "SWMM_online.out")
results <- swmmr::read_out(file = "SWMM_online.out", iType = 1, object_name = "10058", vIndex = 4)

Please let me know what is your output.

However, if we assume you do not get crash:

  • how can I check the RAM memory in my EC2 instance? how can I clear it or do something to make it works?